Recent estimates indicate that only about 40 percent of breast cancer cases can be explained by accepted and suspected risk factors. Exposures to chemical and physical agents in the workplace of women have not been studied adequately. Thus, the object of the present protocol is to determine whether occupational exposures to selected chemical and physical agents is associated with increased risks of pre- and postmenopausal breast cancer. A population-based, case-control study of pre-and postmenopausal breast cancer incidence among women aged 40 to 79 years was conducted in western New York State between 1986 and 1991. Only new, histologically-confirmed primary cases of malignant breast cancer from all major hospitals in the region were enrolled (301 premenopausal and 439 postmenopausal). A population-based control group was recruited and was frequency-matched to case subjects on age and county of residence. Face-to-face interviews by trained nurse- interviewers were carried out and information on a wide array of accepted and potential risk factors was obtained. Standardized lifetime occupational histories were also obtained and these included details regarding place of employment, duties and activities associated with each job. For all 1,550 subjects, the job histories will be sent to a team of expert industrial hygienists and chemists at the Institut Armand-Frappier who will then attribute exposure to about 300 agents using a method developed by Siemiatycki and colleagues. The team has considerable experience in coding these types of job histories and will collaborate with hygienists in Buffalo. The occupational information present in the job descriptions is sufficiently detailed to capture the idiosyncracies of each person's job; thus, this system will minimize exposure misclassification. Statistical analyses, including logistic regression and generalized additive modeling techniques, will be carried out to assess the role of occupational exposures with a focus on polycyclic aromatic hydrocarbons, specific organic solvents and electromagnetic fields. Analyses will be conducted separately for the pre- and postmenopausal groups and will be combined where appropriate. Identification of agents whose exposures are amenable to modification may allow for new public health initiatives to reduce risk. This timely investigation should make an important contribution to our limited understanding of occupational risk factors for breast cancer.